The Snack-Time Speech Practice That Actually Works

For littleWords app, the goal is not to turn parents into therapists. The goal is to make everyday moments easier to join, easier to repeat, and easier for a child to use in their own way.
When Megan, a project manager in Raleigh, got the speech delay evaluation results for her son Caleb at 22 months, the developmental pediatrician handed her a folder. Inside was a photocopied sheet titled “Daily Home Practice Tips.” It listed things like “set aside 15 minutes of dedicated play in a quiet room” and “use intentional language-building toys.” She tried it for three days. On day one, her older daughter kept barging in. On day two, Caleb screamed because the quiet room had no windows and he hated it. On day three, she forgot entirely until 9 p.m. “I felt like I’d already failed,” she told me. “And we were three days in.”
I think most parents of speech-delayed kids have lived some version of Megan’s three days. The advice is technically correct. It is also practically useless for people who live in actual houses with actual schedules.
Here’s the thing: the best speech practice I’ve ever seen didn’t require a quiet room, special toys, or an extra 15 minutes carved out of an already-crumbling routine. It used a snack. That’s it. A snack, a table, and a parent who was already going to be standing in the kitchen anyway.
We did this with my son for two years. I’m going to walk you through exactly how.
A Plate, a Chair, and No Screens
The setup is almost insultingly simple:
- A snack your kid likes
- A small portion on a plate (not a fistful dumped from the bag)
- A cup of water nearby
- One parent sitting across from them, not next to them
- No screens, no toys, just the table
The “across” part is sneakily important. Face-to-face positioning is where eye contact happens naturally. Sitting side by side, you’re both staring at the same wall. Research on shared attention in early language acquisition backs this up: a 2014 study published in Developmental Science found that moments of sustained mutual gaze between parent and child at 11 to 12 months predicted vocabulary size at 18 months (Yu & Smith, 2014). You don’t need to force eye contact. You just need to sit where it can happen on its own.
The small portion matters too. If you put the whole bag on the table, your kid grabs a fistful and wanders off. A plate with three crackers creates a natural transaction loop: they eat a cracker, they need another cracker, they have a reason to communicate with you. A 2009 analysis in the Journal of Speech, Language, and Hearing Research found that naturalistic communication opportunities embedded in daily routines like meals were as effective as structured therapy activities for increasing word production in toddlers with language delays (Roberts & Kaiser, 2011). The plate is your structure. The cracker is your prompt.
We ran this every weekday between my son’s two-year evaluation and his fourth birthday. By the end, the routine had become invisible. He’d walk to the table at 3 p.m. and look at me expectantly, like a tiny colleague arriving at a standing meeting. Let’s do this.
Four Moves, Repeated Until the Crackers Run Out
The whole technique is four steps on a loop:
Ask. “Do you want crackers?” or “Apples or pretzels?” or just “More?”
Wait. Five full seconds. Count them in your head. This is the hardest part for parents. We are pathologically uncomfortable with silence. But silence is the gap where language tries to climb through. Speech-language pathologists call this “expectant waiting” or “time delay,” and research from Halle, Marshall, and Spradlin (1979) showed it to be one of the most effective low-effort strategies for prompting spontaneous communication in children with delays. Five seconds feels like an eternity when you’re staring at a toddler who is staring at a cracker. Do it anyway.
Model. If nothing comes, say the word you’re hoping for. “Crackers. You want crackers.” Put the crackers down. You are not correcting them. You are not testing them. You are simply showing them what the word sounds like, right at the moment they care about the thing the word describes.
Expand. If he says “cackers,” you say “crackers, yes! Crunchy crackers.” You expand by one word, maybe two. You do not deliver a TED Talk about the nutritional content of whole grains. Researchers call this “recasting” and “expansion,” and a 2006 meta-analysis published in the American Journal of Speech-Language Pathology confirmed that these strategies produce measurable gains in expressive language when used consistently over time (Cleave et al., 2015).
Ask, wait, model, expand. Over and over. Through the whole snack. Ten minutes, give or take. That’s your session.
What Actually Changed Over Twelve Months
Nobody tells you that the same routine will look completely different at month one, month six, and month twelve. The structure stays constant. The kid is the variable.
Month one (no words). Most of snack time was me narrating to a kid who was chewing and looking at the wall. He pointed at the crackers. I said “cracker” and handed him one. I said “more?” before each one. He heard maybe 50 words of relevant vocabulary in 10 minutes. That was the entire goal. It felt like talking to a polite houseplant, and I had to keep reminding myself it was working. A 2003 study by Hart and Risley (frequently cited as the foundational “30 Million Word Gap” research) found that the sheer quantity of words children hear in their early years directly correlates with language outcomes. Even if your kid produces nothing, the input is building something you cannot see yet.
Month three. He started saying “ca” for cracker. I celebrated. I modeled the full word back. Sometimes I added a descriptor. “Crunchy cracker!” He copied me occasionally. He chewed most of the time. Still 10 minutes. Still calm. The partial word was a huge signal. It meant he had the concept, the intent to communicate, and enough motor planning to attempt the sound. SLPs call these “phonological approximations,” and they are not failures. They are the scaffolding of real words.
Month six. “More ca” with a point. Two-word combinations arriving. I expanded: “More crackers, yes!” I started adding choice questions. “Crackers or pretzels?” He’d pick. Sometimes he used the word. Sometimes he just pointed. Both fine. The transition from single words to two-word combinations is a well-documented milestone in typical language development, usually occurring between 18 and 24 months. When it arrives late, it tends to accelerate once it starts, especially if the child is getting consistent input (Rescorla, 2011).
Month nine. “I want more crackers please.” A full sentence, spoken at the kitchen table over Goldfish, and I had to pretend I wasn’t about to cry. Snack time was becoming actual conversation. I asked about his day. He told me a halting story about preschool. The jump from two-word combinations to full sentences often happens fast once a child’s vocabulary crosses the 50-word threshold, a phenomenon sometimes called the “vocabulary burst.” For late talkers, this burst can feel sudden and dramatic after months of slow accumulation.
Month twelve. Snack time was just snack time with a kid who talked. The scaffolding had turned invisible. He didn’t need the prompts. He initiated. He asked me questions. He told jokes (bad ones, but still).
This worked because we did it every weekday for two years. There is no hack for consistency. Consistency is the hack.
Small Variations That Kept Me From Losing My Mind
Doing the exact same thing 250 times a year will hollow you out if you don’t mix it up a little. We rotated loosely:
Color days. “What color is this cracker?” “What other things are orange?” “Tell me an orange thing in this room.”
Shape days. “Is the cracker round or square?” “What else is square?”
Texture days. “Is the cracker hard or soft?” “What about the apple?” Sensory adjectives are especially useful because your kid is literally experiencing the texture in their mouth at the moment you name it. That simultaneity of experience and language is exactly what makes mealtime practice so effective compared to flashcard drills.
Story days. “Who do you want to tell about your snack? Let’s pretend to call grandma.” (Hand him a banana as a phone.) “Hi grandma, I’m eating crackers.” This one turned out to be his favorite. Narrative play, even at its simplest, builds sentence structure and sequencing skills that matter for later literacy (Peterson & McCabe, 1983).
Choice days. Lay out three options. Make him pick. Make him use a word. As his vocabulary grew, we added a follow-up: “Why did you pick the pretzels?” Even a one-word answer (“salty”) practices reasoning and descriptive language.
You don’t need all of these. You don’t need any particular one on any particular day. The variations aren’t for your kid. They’re for you. They keep you from burning out on a routine that, honestly, can feel monotonous if you let it.
The Days It Fell Apart
Plenty of sessions were disasters. Some days he wasn’t hungry. Some days he was so overstimulated from preschool that sitting at a table felt like asking him to meditate in a thunderstorm. Some days I was the one who couldn’t focus, scrolling my phone under the table, and he picked up on it instantly. Kids always pick up on it.
There were also days that fell apart in more specific ways. One afternoon he threw the plate because I offered bananas instead of crackers. Another time he just said “no” to every question for six straight minutes, which, looking back, was its own kind of language practice but didn’t feel like a win at the time. On at least two occasions I sat down and realized I had nothing prepped, so we ate dry cereal out of a measuring cup. It still worked. The food is the vehicle, not the point.
The rule we developed: don’t force it. If today isn’t working, just hand him the snack and try again tomorrow. The compound interest of showing up most days beats the one-time heroics of muscling through a bad session.
We skipped or shortened roughly 30 percent of our sessions. The other 70 percent did the work. Think of it like going to the gym: nobody goes every single day, and the people who insist they will are the ones who quit by February.
Adding a Second Touchpoint
Around month nine, when his language had real momentum, our SLP suggested adding an evening conversational practice in addition to snack time. The logic: he needed more language exposure than meals could provide, and I needed something that didn’t require me to be “on” for another 10 minutes.
We added the LittleWords app for a 10-minute window after his sister went to sleep. He’d lie on the couch with the tablet, talk to Buddy (the AI character) about his day, and decompress.
It was a good pairing. Snack time was high-attention, high-eye-contact, parent-led. The app was low-pressure, kid-led, conversational. He needed both modes. The app did not replace snack-time practice. It complemented it, the way stretching complements running but doesn’t substitute for it.
I want to be specific about what the app did well in our case. It gave him a no-stakes conversational partner at a time of day when I was running on fumes. He could repeat himself, stumble over words, and try again without anyone’s facial expression shifting. For a kid who was starting to notice that he talked differently from his peers, that low-pressure space mattered. It also gave me a break from being the only person creating language opportunities, which, after nine months of daily snack sessions, I needed badly.
By the end of the year, his SLP said he’d made about 18 months of progress in 12 calendar months. I don’t credit any single thing for that. I think it was the layering: structured mealtime conversation, plus app-based practice, plus weekly therapy. Multiple exposures across the day, none of them requiring heroic effort on their own.
Your Seven-Day Starter
If you want to try this, here’s one week. Just one. See what happens.
Day one. Set up the table. Ask, wait, model. Five minutes. Don’t worry about whether your kid talks. Just establish the rhythm.
Day two. Add a choice. “Crackers or pretzels?” Wait. Hand them whichever they choose.
Day three. Add the “more?” question. Accept a sign, a word, a vocalization. Anything counts.
Day four. Add a descriptor. “Crunchy cracker.” “Sweet apple.” Don’t quiz. Just narrate.
Day five. Try a story. “Tell me about your day.” Even if you get one word back, expand it. “You played outside? With friends?”
Day six. Repeat day five. Let it feel like conversation, not homework.
Day seven. Sit at the table and see what your kid does. By now, the pattern is in their body. They might surprise you.
Two years from now, you’ll have a kid who associates eating with talking. That is, honestly, the single best gift you can give a language-delayed child. Food brings them to the table. Your presence keeps them there. The words come.
FAQ
How old does my child need to be for this to work? We started at 24 months, but the technique works for any child old enough to sit in a high chair or booster seat and eat finger foods, typically around 12 months. For very young toddlers, you are just modeling and narrating. The “wait” step still applies; even pre-verbal children benefit from the pause because it teaches them that communication has a turn-taking structure. If your child is older, say three or four, you simply adjust the complexity of your questions and expansions.
What if my child uses a speech-generating device or AAC system? The snack-time structure works well with augmentative and alternative communication. Instead of waiting for a spoken word, you wait for them to activate their device or point to a symbol. The same loop applies: ask, wait for them to respond via their device, model the spoken word alongside the device output, and expand. Many SLPs recommend pairing AAC with naturalistic routines exactly like this one (Romski & Sevcik, 2005).
Does the type of snack matter? Not nutritionally, for these purposes. What matters is that the snack is something your child actually wants enough to stay at the table for, and that it comes in small, discrete pieces so you can create multiple request opportunities. Goldfish crackers, blueberries, cheese cubes, pretzel sticks, and small apple slices all work well. Pouches and squeezable foods are less ideal because they don’t create the natural pause between bites where communication happens.
What if I have two kids who both need this? Run separate sessions if you can. The one-on-one attention is a big part of why this works. Even a five-minute solo snack with each child is better than a fifteen-minute group session where neither child gets enough wait time. If separate sessions are genuinely impossible, alternate who you direct questions toward and make sure each child gets their own full five-second pause.
Should I stop doing this once my child catches up? You’ll stop naturally. That’s the beauty of embedding speech practice in a routine you were already doing. When your child starts talking freely at the table, snack time just becomes snack time. You don’t need to formally discontinue it. You’ll realize one afternoon that you’ve been having a normal conversation for ten minutes without thinking about modeling or expanding, and that’s when you know.
Can this replace speech therapy? No. This is a home supplement, not a clinical intervention. Your SLP provides assessment, targeted exercises for specific speech sound errors or language deficits, and professional monitoring of progress. What snack-time practice does is give your child dozens of additional naturalistic language opportunities per week in a context that feels normal rather than clinical. The two work together. If your child has been evaluated and recommended for therapy, follow through with that.
My partner and I have different communication styles. Does that matter? It can actually help. Research on language input diversity suggests that hearing varied conversational styles from multiple caregivers broadens a child’s pragmatic language skills (Hoff, 2006). If one parent asks more questions and the other narrates more, your child gets exposure to both patterns. The key is that both caregivers understand the basic loop (ask, wait, model, expand) and commit to the wait step. The most common mismatch I’ve seen is when one parent waits the full five seconds and the other jumps in after one. Talk about the timing together before you start.
This article describes our family’s experience with home-based speech practice strategies. It is not medical or clinical advice. Every child’s speech and language needs are different. Please consult a licensed speech-language pathologist for guidance specific to your child.